PROJECT SUMMARY Five years of research in the Emory ACE have provided compelling evidence for a causal link between early deficits in social engagement and later deficits in speech and language in infants at risk of autism spectrum disorder (ASD) across every stage of vocal development, beginning in the first months of life. Developmental progressions in vocal behavior are found not only in the infant, but also in the caregiver, and progressions in both infant and caregiver are disrupted by autism, suggesting that detection and intervention targeting spoken communication in ASD should be grounded in measuring and manipulating social contingency in dyadic interactions early in life, attending to the mother as well as the child. Extending our research to explore these findings, Project II will identify developmental progressions in social contingency between infant and caregiver, discover active ingredients and developmental windows of opportunity for key mechanisms of social interaction that scaffold early vocal development and the emergence of speech and language, and determine how these differ in autism and typical development. A new cohort of 150 high-risk infants and 100 low-risk controls from 0 to 30 months will allow us to analyze effects of sex, race and ethnicity, and socioeconomic status. The first specific aim is to identify pivotal transitions in vocal behavior and social contingency in both infant and caregiver over the first three years of life. Based on automated analysis of home audio recordings sampled longitudinally at monthly intervals, we map out developmental progressions in patterns of infant-directed and adult-directed vocal behavior, and identify periods where there are significant changes in typical development and delays or deviations in autism. The second specific aim is to identify mechanisms responsible for pivotal transitions, in infant and caregiver. By measuring differences in acoustic cues and timing patterns in infant/adult-directed speech over development, we identify active ingredients of vocal signaling that mediate transitions, and test whether infants/caregivers are producing/responding to signals appropriately. The third specific aim is to evaluate the performance of developmental trajectories of social contingency and vocal behavior as biomarkers of risk and predictors of treatment response and outcome. Combining results across Projects I-IV, we test the hypothesis that measures of early social contingency are more effective at predicting speech and language outcome in infants at risk of ASD than measures of early vocal production. We predict that treatment paradigms targeting social contingency between infant and caregiver are more effective at promoting speech development than those that target speech development in the infant alone. This research addresses all four aims of the NIMH Strategic Plan.